Summary Organoid models have recently emerged as an alternative, more realistic in vitro representation of patient tumors when compared to traditional 2D culture approaches. At a functional level, organoids have proven useful as realistic tumor model to be used for ex vivo drug testing. Although organoid models can often accurately recapitulate the spatial organization of the tumor, they are typically derived from a single cell type (e.g., tumor cells or pluripotent stem cells) and existing tools for making organoids with well-defined compositions are limited. To address this, we propose to develop an instrument for building organoid co-cultures consisting of the major tumor, stromal, and immune cell types previously shown to play a role in drug resistance in multiple myeloma, by fluorescently labeling cells and dispensing them from patient samples into culture wells in a single step. By monitoring the growth and viability of these organoids and subjecting them to drug testing, we will determine the minimal functional unit that is sufficient to recapitulate the major tumor-stroma interactions of the multiple myeloma bone marrow niche. If successful, we envision that more realistic patient-derived organoid models could lead to improved functional assays to test the susceptibility of patient samples to cancer therapeutics, with the potential goal of using these assays to guide treatment decisions.